Filters








34,498 Hits in 8.7 sec

Improving Visualization Of High-Dimensional Music Similarity Spaces

Arthur Flexer
2015 Zenodo  
as a new aspect of the curse of dimension- high dimensional music similarity spaces.  ...  “Improving visual- For our experiments we used two standard music databases: ization of high-dimensional music similarity spaces”, 16th International the “GTZAN” collection consisting of N  ... 
doi:10.5281/zenodo.1416472 fatcat:2ibks2zq2zghpd5fl2jhl6r2m4

Visually and Acoustically Exploring the High-Dimensional Space of Music

Lukas Bossard, Michael Kuhn, Roger Wattenhofer
2009 2009 International Conference on Computational Science and Engineering  
In this paper we discuss two alternative exploration schemes, both taking advantage of a recently proposed high-dimensional music similarity space.  ...  The permanent growth of personal music collections caused by the ongoing digital revolution asks for novel ways of organization.  ...  By decomposing the high-dimensional music similarity space dependent on the given feedback, it is able to identify regions of interest of a given user, can thus constantly improve its estimate of the user's  ... 
doi:10.1109/cse.2009.131 dblp:conf/cse/BossardKW09 fatcat:gwvsmowvhzbibklbrh6kvxfx54

Soundanchoring: Content-Based Exploration Of Music Collections With Anchored Self-Organized Maps

Leandro Collares, Tiago Fernandes Tavares, Joseph Feliciano, Shelley Gao, George Tzanetakis, Amy Gooch
2013 Proceedings of the SMC Conferences  
This dimensionality reduction technique preserves the topology Proceedings of the Sound and Music Computing Conference 2013, SMC 2013, Stockholm, Sweden of the high-dimensional space as much as possible  ...  The set of feature vectors is a high-dimensional space that is mapped to two dimensions using the AnchoredSOM algorithm.  ... 
doi:10.5281/zenodo.850401 fatcat:txx36fqfrrbf3phrbozmg53fqu

Using Smoothed Data Histograms for Cluster Visualization in Self-Organizing Maps [chapter]

Elias Pampalk, Andreas Rauber, Dieter Merkl
2002 Lecture Notes in Computer Science  
Several methods to visualize clusters in high-dimensional data sets using the Self-Organizing Map (SOM) have been proposed.  ...  The method is illustrated using a simple 2-dimensional data set and similarities to other SOM based visualizations and to the posterior probability distribution of the Generative Topographic Mapping are  ...  A model vector in the high-dimensional data space is assigned to each of the units.  ... 
doi:10.1007/3-540-46084-5_141 fatcat:5b2tiydfivcfdi5khjylbp6swi

Streamcatcher: Integrated Visualization Of Music Clips And Online Audio Streams

Martin Gasser, Arthur Flexer, Gerhard Widmer
2008 Zenodo  
By reducing dimensionality of the feature space with a Principal Component Analysis, he also mapped audio signals from high-dimensional timbresimilarity space into a three-dimensional visualization space  ...  Acoustic similarity is mapped to proximity data in a 2D visualization, which in turn is derived from a high dimensional timbre similarity space by means of multidimensional scaling.  ... 
doi:10.5281/zenodo.1416361 fatcat:xelxwnowzrdflpf5xqtyw4dehm

From Web to Map: Exploring the World of Music

Olga Goussevskaia, Michael Kuhn, Michael Lorenzi, Roger Wattenhofer
2008 2008 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology  
In this work we propose to use a high-dimensional map of the "world of music" as a data structure for music retrieval and exploration of personal collections.  ...  As a concrete example, we have developed a web-application that allows users to visualize and navigate through their music collections and create playlists by specifying trajectories.  ...  space to improve the quality of the similarity measure.  ... 
doi:10.1109/wiiat.2008.20 dblp:conf/webi/GoussevskaiaKLW08 fatcat:w7vuezhfanatff3nvdd5mv2day

Databionic Visualization Of Music Collections According To Perceptual Distance

Fabian Mörchen, Alfred Ultsch, Mario Nöcker, Christian Stamm
2005 Zenodo  
ACKNOWLEDGEMENTS The authors would like to thank Ingo Löhken, Michael Thies, Niko Efthymiou, and Martin Kümmerer for fruitful discussion on this research and for the development of the MusicMiner.  ...  It is rather a pixel in a high resolution display of the projection from the high dimensional data space to the low dimensional map space.  ...  The visualizations based on the paradigm of topographic maps enables an intuitive navigation of the high dimensional feature space.  ... 
doi:10.5281/zenodo.1417966 fatcat:bjwojcmpmfcprf6m4xgr5ryuzm

Nonlinear dimensionality reduction approaches applied to music and textural sounds

Stephane Dupont, Thierry Ravet, Cecile Picard-Limpens, Christian Frisson
2013 2013 IEEE International Conference on Multimedia and Expo (ICME)  
Although PCA and ISOMAP can yield good continuity performance even locally (samples in the original space remain close-by in the low-dimensional one), they fail to preserve the structure of the data well  ...  ISOMAP and t-SNE are being compared to PCA in a visualization problem, where we end up with a two-dimensional view.  ...  It does so using a cost function that favors the probability distributions of points belonging to the neighborhoods of other points to be similar in the high-dimensional space and in its low-dimensional  ... 
doi:10.1109/icme.2013.6607550 dblp:conf/icmcs/DupontRPF13 fatcat:4lrorgh2x5fqpei2537iaudal4

geMsearch: Personalized Explorative Music Search

Christian Esswein, Markus Schedl, Eva Zangerle
2018 International Conference on Intelligent User Interfaces  
This allows for efficient approximate querying of the collection and, more importantly, for employing visualization strategies that allow the user to explore the music collection in a 3D-space.  ...  Due to the rise of music streaming platforms, huge collections of music are now available to users on various devices.  ...  of the suggested items, allowing users to visually explore the music collection in a 3D-space.  ... 
dblp:conf/iui/EssweinSZ18 fatcat:acqeolrgoze2vah7j45ehffb6q

A Comparative Study On Improving Sound Similarity Maps With Semantic Metadata

Lennart Nicolas Krebs, Frederic Font, Henrik Hahn
2022 Zenodo  
These maps usually rely on dimensionality reduction methods like PCA, UMAP, or t-SNE to translate an audio embedding or another high dimensional feature representation into a low dimensional latent space  ...  We conducted a comparative study of different map layouts to understand the usefulness of the aforementioned method to improve sound similarity projections.  ...  to improve sound searching and organization with visual approaches like two-dimensional sound maps.  ... 
doi:10.5281/zenodo.7116193 fatcat:oaumtusz55cajnnyid3dvjijfq

Visualizing changes in the structure of data for exploratory feature selection

Elias Pampalk, Werner Goebl, Gerhard Widmer
2003 Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '03  
Using visualization techniques to explore and understand high-dimensional data is an efficient way to combine human intelligence with the immense brute force computation power available nowadays.  ...  We demonstrate the application of our approach in two music related data mining projects.  ...  This research has been carried out in the project Y99-INF, sponsored by the Austrian Federal Ministry of Education, Science and Culture (BMBWK) in the form of a START Research Prize.  ... 
doi:10.1145/956755.956771 fatcat:cfjtlur7pfdqtabacg23pys274

Visualizing changes in the structure of data for exploratory feature selection

Elias Pampalk, Werner Goebl, Gerhard Widmer
2003 Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '03  
Using visualization techniques to explore and understand high-dimensional data is an efficient way to combine human intelligence with the immense brute force computation power available nowadays.  ...  We demonstrate the application of our approach in two music related data mining projects.  ...  This research has been carried out in the project Y99-INF, sponsored by the Austrian Federal Ministry of Education, Science and Culture (BMBWK) in the form of a START Research Prize.  ... 
doi:10.1145/956750.956771 dblp:conf/kdd/PampalkGW03 fatcat:m6jinqe2wvgibjbde7osujhkx4

Visualization and Clustering of Tagged Music Data [chapter]

Pascal Lehwark, Sebastian Risi, Alfred Ultsch
2008 Studies in Classification, Data Analysis, and Knowledge Organization  
The usage of Emergent-Self-Organizing-Maps (ESOM) and U-Map techniques to visualize and cluster this sort of tagged data to discover emergent structures in collections of music is reported.  ...  This kind of user generated content can be used to define a similarity measure for those objects.  ...  high dimensional data.  ... 
doi:10.1007/978-3-540-78246-9_79 fatcat:k7n5aeudzzcaxddvhcldsassyq

SmartDJ: An interactive music player for music discovery by similarity comparison

Maureen S. Y. Aw, Chung Sion Lim, Andy W. H. Khong
2013 2013 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference  
This prototype application enables users to visualize their music library, select songs based on their similarity or automate the song selection process using a given seed song.  ...  The similarity between songs is given by the Euclidean distance in this lower-dimension song space.  ...  The authors would like to thank PerMagnus Lindborg from the School of Art, Media and Design, Nanyang Technological University for his insights into music mixing.  ... 
doi:10.1109/apsipa.2013.6694280 dblp:conf/apsipa/AwLK13 fatcat:vmzjhw2hczcafjodf62t2vmzzy

Music Thumbnailer: Visualizing Musical Pieces In Thumbnail Images Based On Acoustic Features

Kazuyoshi Yoshii, Masataka Goto
2008 Zenodo  
Therefore, our objective was to find an appropriate mapping from a low-dimensional acoustic space (several-tens dim.) to a high-dimensional visual space (several-thousands dim.).  ...  This transformation has a high degree of freedom; a higher-dimensional space can always preserve the complete information of an original space.  ... 
doi:10.5281/zenodo.1415936 fatcat:shtfxpo5ebgqvb733n6375isl4
« Previous Showing results 1 — 15 out of 34,498 results